mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-13 15:38:22 +01:00
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - The chat-feature evals measure whether agents complete real chat requests through the product > - The chat catalog eval harness and its repairs live on the `chat-eval/master-candidate` branch, not on master > - Today's work made 139 of 242 chat evals pass in verified single-case live runs. Each product fix landed on master as its own pull request > - The eval-side changes (setups, graders, fault seams, evidence capture) still exist only on the candidate > - This pull request lands the harness on master. Master stays the source of truth for product behavior > - The benefit is one tree: evals run against master commits, and product fixes reach the evals as soon as they merge ## Linked Issues or Issue Description Refs #19441, Refs #19599, Refs #20270, Refs #20334, Refs #20342, Refs #20344, Refs #20354, Refs #20360, Refs #20373, Refs #20382, Refs #20409 **What existing behavior does this improve?** The chat catalog eval harness and the product it evaluates live on a long-running branch outside master. Eval fixes do not reach master, so master-based eval runs use old setups and graders. **Subsystem affected** `tests/runner-e2e` chat catalog harness, plus a small set of product changes that exist only on the candidate. **Proposed change** Land the harness on master. Every product fix from this eval work already merged to master as its own pull request. Conflicts keep master's product behavior. ## What Changed - Most of the diff (808 of 851 files) is under `tests/runner-e2e/`: chat-catalog domains, scenarios, descriptors, the TestOps manifest, private evidence tooling, and their tests. - Graders now check saved state, setup literals and rendered or clicked links. Many keyword and regex checks on free-form replies are removed. Each changed grader has a bumped `oracleVersion` and calibration tests (a valid reply, a reworded valid reply, and broken variants that must fail). - The shared link helper `hasLink` no longer treats a list link (for example `/COM/issues?view=mine`) as a link to every task without a URL key. - Setup fixes build real state through supported product routes: routine history, agent grants through the board grant route, budget holds with a real cap, and a storage-layer archive fault for `inbox-search.archive-partial-failure`. - Browser boundary journeys (sign-in, chat-disabled, onboarding) now write the standard pass evidence files. - Retargeted cases, approved by the maintainer: `agent-chat.chat-disabled` (Conference Room gate), `agents.restore-instructions` (no saved instruction history), `access.manage-membership` (member permission edits stay board-only). - The private evidence trusted files (`manual-private-handoff.ts`, `manual-private-retention.ts`, `manual-private-report-assets.ts`) are byte-identical to master. - Product files outside `tests/` (43 files, +1,328 −49) carry candidate-only changes that master does not yet have. Most are tests. The main code changes are in `request-human-input.ts`, `heartbeat-run-summary.ts`, `native-completion-feedback.ts`, `native-run-finalizer.ts`, `openapi.ts`, `secrets.ts` and `IssueChatLongThreadPerf.tsx`. ## Verification - `pnpm vitest run chat-catalog-domains chat-catalog-contract.test.ts chat-catalog-consistency.test.ts unit-launcher.test.mjs --testTimeout=180000` in `tests/runner-e2e`: 269 files pass, 3,550 tests pass. - `node --test chat-full-catalog-testops-manifest.test.mjs`: 10 pass. - `node tests/runner-e2e/source-sdk-public-inputs.mjs <HEAD> --check`: no change. - `tsc` for `server` and `tests/runner-e2e`: clean. - Live single-case verification: 139 of 242 cases passed on paid runs, and a person read the saved evidence for each. These are single-case runs, not one full-cohort run. ## Risks - Large diff. Most of it is test-only, but the 43 non-test files change product code paths. Review those files first. - Some graders are now less strict on reply wording. A person must read replies for meaning on those cases. - 42 cases need disposable outside accounts (Notion, Slack, GitHub, Daytona and others) and stay blocked. `access.invite-teammate` stays failing until local agent isolation lands, because the runner's durable state keeps a live invite token. - The paid workflow compares the trusted private-evidence files with master. They are unchanged, so this risk is low. ## Model Used Anthropic Claude Opus 5.5 (`claude-opus-5-5`), with tool use and code execution, in Claude Code. Subagents used the same model. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] All Paperclip CI gates are green - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Paperclip <noreply@paperclip.ing>
90 lines
7.8 KiB
TypeScript
90 lines
7.8 KiB
TypeScript
import { afterEach, describe, expect, it, vi } from "vitest";
|
|
import type { ChatCatalogContext } from "./chat-catalog-contract.js";
|
|
import { chatCatalogNativePhaseDeadline, chatCatalogNativePhaseIsTimely, chatCatalogNativePhaseRemainingBudget, chatCatalogOriginalPhaseDeadline } from "./chat-catalog-contract.js";
|
|
const observed = vi.hoisted(() => vi.fn());
|
|
vi.mock("./chat-catalog-domains/internal-support.js", async importOriginal => ({ ...await importOriginal<object>(), observeInternal: observed }));
|
|
import { waitForReply } from "./chat-catalog-domains/identity-support.js";
|
|
import { settlePrivilegedChat } from "./chat-catalog-domains/privileged-support.js";
|
|
function context(scenarioId = "cli-operations.owned-dev-instance", deadline = 870_000): ChatCatalogContext {
|
|
return { scenario: { id: scenarioId }, input: { execution: { id: `chat-features.runner-opencode.local.${scenarioId}` } }, observationDeadlineAt: deadline,
|
|
issue: () => ({ conversationState: "waiting" }) } as unknown as ChatCatalogContext;
|
|
}
|
|
afterEach(() => { vi.useRealTimers(); vi.restoreAllMocks(); vi.clearAllMocks(); });
|
|
describe("selected native phases share the once-created original flow clock", () => {
|
|
it.each(["secrets.secret-proposal", "developer-labs.long-thread-performance", "cli-operations.owned-dev-instance", "recovery.controller-restart", "routines.run-now", "task-coordination.independent-review", "steering.edit-queued-message", "tasks.website-without-project", "inbox-search.archive-partial-failure", "questions-and-approvals.approve-tool-once", "runs-adapters.reset-session"])("%s uses absolute remaining outer budget rather than resetting its shorter cap", scenario => {
|
|
vi.useFakeTimers(); vi.setSystemTime(800_000); const ctx = context(scenario);
|
|
expect(chatCatalogNativePhaseDeadline(ctx, Date.now() + 150_000)).toBe(870_000);
|
|
vi.setSystemTime(850_000); expect(chatCatalogNativePhaseDeadline(ctx, Date.now() + 180_000)).toBe(870_000);
|
|
expect(ctx.observationDeadlineAt).toBe(870_000);
|
|
});
|
|
it.each(["instance-operations.backup-and-restore", "instance-operations.restart-service", "routines.webhook-trigger", "status-cards.explain-card", "secrets.configure-company-secret", "teams.configure-team", "steering.interrupt-with-direction", "companies.fast-responses"])("future13 %s shares the setup-inclusive original clock", scenario => {
|
|
vi.useFakeTimers(); vi.setSystemTime(800_000); const ctx = context(scenario);
|
|
expect(chatCatalogNativePhaseDeadline(ctx, Date.now() + 150_000)).toBe(870_000);
|
|
expect(chatCatalogOriginalPhaseDeadline(ctx)).toBe(870_000);
|
|
vi.setSystemTime(869_999); expect(chatCatalogNativePhaseRemainingBudget(ctx)!()).toBe(1);
|
|
vi.setSystemTime(870_000); expect(() => chatCatalogOriginalPhaseDeadline(ctx)).toThrow(/exhausted/);
|
|
});
|
|
it.each(["secrets.configure-company-secret", "companies.fast-responses"])("actual %s identity helper settles after its old150s cap using only the original remaining clock", async scenario => {
|
|
vi.useFakeTimers(); vi.setSystemTime(50_000); const originalDeadline = scenario === "companies.fast-responses" ? 390_000 : 450_000; const ctx = context(scenario, originalDeadline);
|
|
Object.assign(ctx, { refreshIssue: async () => {}, allRuns: async () => [{ status: Date.now() >= 230_000 ? "succeeded" : "running" }] });
|
|
let completed = false; const pending = waitForReply(ctx, chatCatalogOriginalPhaseDeadline(ctx)).then(() => { completed = true; });
|
|
await vi.advanceTimersByTimeAsync(152_000); expect(completed).toBe(false);
|
|
await vi.advanceTimersByTimeAsync(28_000); await pending; expect(completed).toBe(true);
|
|
expect(ctx.observationDeadlineAt).toBe(originalDeadline);
|
|
});
|
|
it.each(["onboarding.first-work", "routines.plugin-managed-routine", "status-cards.archive-card", "status-cards.other-authors-card", "companies.prompt-injection-guard", "companies.configure-decision-model"])("future13 %s does not receive new phase or UI behavior", scenario => {
|
|
vi.useFakeTimers(); vi.setSystemTime(800_000); const ctx = context(scenario, 1);
|
|
expect(chatCatalogNativePhaseDeadline(ctx, 900_000)).toBe(900_000);
|
|
expect(chatCatalogOriginalPhaseDeadline(ctx)).toBeUndefined();
|
|
expect(chatCatalogNativePhaseRemainingBudget(ctx)).toBeUndefined();
|
|
});
|
|
it("actual existing privileged phase can settle beyond150s without changing native/reply proof", async () => {
|
|
vi.useFakeTimers(); vi.setSystemTime(0); const ctx = context();
|
|
observed.mockImplementation(async () => ({ reply: Date.now() >= 180_000 ? "Use the operator path." : "", runs: [{ status: Date.now() >= 180_000 ? "succeeded" : "running" }] }));
|
|
let complete = false; const pending = settlePrivilegedChat(ctx).then(() => { complete = true; });
|
|
await vi.advanceTimersByTimeAsync(152_000); expect(complete).toBe(false);
|
|
await vi.advanceTimersByTimeAsync(28_000); await pending; expect(complete).toBe(true);
|
|
});
|
|
it("late successful reads, running natives and extra queued rows cannot pass at the deadline", async () => {
|
|
vi.useFakeTimers(); vi.setSystemTime(869_000); const ctx = context();
|
|
observed.mockImplementation(async () => { vi.setSystemTime(871_000); return { reply: "Use settings", runs: [{ status: "succeeded" }] }; });
|
|
const late = settlePrivilegedChat(ctx).then(() => "accepted", () => "expired");
|
|
await vi.advanceTimersByTimeAsync(3000); expect(await late).toBe("expired");
|
|
for (const runs of [[{ status: "running" }], [{ status: "succeeded" }, { status: "queued" }]]) {
|
|
vi.setSystemTime(869_000); observed.mockResolvedValue({ reply: "Use settings", runs });
|
|
const pending = settlePrivilegedChat(ctx).then(() => "accepted", () => "expired");
|
|
await vi.advanceTimersByTimeAsync(3000); expect(await pending).toBe("expired");
|
|
}
|
|
});
|
|
it("fails closed on missing/expired/nonfinite clocks and never manufactures grace", () => {
|
|
vi.useFakeTimers(); vi.setSystemTime(870_000);
|
|
for (const deadline of [undefined, NaN, Infinity, 870_000, 869_999]) {
|
|
const ctx = context(); Object.assign(ctx, { observationDeadlineAt: deadline });
|
|
expect(() => chatCatalogNativePhaseDeadline(ctx, Date.now() + 180_000)).toThrow(/exhausted or missing/);
|
|
expect(chatCatalogNativePhaseIsTimely(ctx)).toBe(false);
|
|
}
|
|
});
|
|
it("declared followups consume the same clock and never return an unlimited zero timeout", () => {
|
|
vi.useFakeTimers(); vi.setSystemTime(869_999); const remaining = chatCatalogNativePhaseRemainingBudget(context())!;
|
|
expect(remaining()).toBe(1); vi.setSystemTime(870_000); expect(remaining).toThrow(/exhausted/);
|
|
expect(chatCatalogNativePhaseRemainingBudget(context("recovery.post-stop-message"))).toBeUndefined();
|
|
});
|
|
it("keeps the one remaining millisecond and rejects final proof at exact expiry", () => {
|
|
vi.useFakeTimers(); vi.setSystemTime(869_999); const ctx = context();
|
|
expect(chatCatalogNativePhaseDeadline(ctx, Date.now() + 150_000) - Date.now()).toBe(1);
|
|
expect(chatCatalogNativePhaseIsTimely(ctx)).toBe(true);
|
|
vi.setSystemTime(870_000); expect(chatCatalogNativePhaseIsTimely(ctx)).toBe(false);
|
|
});
|
|
it.each(["secrets.ask-for-plaintext", "steering.preserve-draft", "routines.read-history", "recovery.post-stop-message", "companies.rename-company"])("%s keeps its existing phase semantics", scenario => {
|
|
vi.useFakeTimers(); vi.setSystemTime(800_000); const ctx = context(scenario, 1);
|
|
expect(chatCatalogNativePhaseDeadline(ctx, 950_000)).toBe(950_000); expect(chatCatalogNativePhaseIsTimely(ctx)).toBe(true);
|
|
});
|
|
it("does not opt another provider/environment lane into the selected scope", () => {
|
|
const ctx = context();
|
|
for (const id of ["chat-features.runner-codex.local.cli-operations.owned-dev-instance", "chat-features.runner-opencode.daytona.cli-operations.owned-dev-instance"]) {
|
|
Object.assign(ctx.input.execution, { id });
|
|
expect(chatCatalogNativePhaseDeadline(ctx, 150_000)).toBe(150_000); expect(chatCatalogNativePhaseIsTimely(ctx)).toBe(true);
|
|
}
|
|
});
|
|
});
|